PROJECT ABSTRACT : Management of localized muscle-invasive bladder cancer (MIBC) with radical cystectomy is associated with risk of significant morbidity, perioperative mortality, and decreased quality of life- especially among the growing number of elderly. A less invasive and morbid, but equally effective means of treatment is needed. Multi-modal treatment of MIBC (maximal trans-urethral resection, followed by combined chemo and radiotherapy) is, for well-selected patients, associated with survival outcomes comparable to cystectomy and lower morbidity and mortality, but also offers bladder preservation. However, in the U.S., multi-modal therapy is not common for 3 reasons: First, there is uncertainty among referring urologists as to whether all tumor is visible at time of resection. Second, no currently available radiographic technologies allow reliable visualization of the bladder tumor site and its margins. Furthermore, the bladder is a deformable, mobile structure, whose location, size and shape vary depending on bladder and bowel fullness. This makes it very possible that the whole bladder and tumor site are not consistently treated, and that collateral radiation occurs. Third, we lack a non-invasive (e.g. imaging-based) means by which to assess for local control after treatment. Emerging data suggests that multi-parametric diffusion- weighted and contrast enhanced MRI can detect occult disease, but this has not been robustly validated because imaging findings, to date, could not be definitively correlated with histology of the same bladder site. We describe a means to reliably mark the bladder margins and tumor site to resolve these limitations of multi-modal therapy. To date, no means of marking the bladder walls has proven effective and reliable. We developed a novel fiducial marker design and placement protocol, initially to simply improve targeting of the tumor-site with high-dose radiotherapy. These ultra-small 24K gold markers are placed endoscopically in submucosa around the tumor site at time of pre-treatment maximal resection, and, at the anatomic boundaries of the bladder, to optimize image-guided bladder radiotherapy. To date, we have placed our markers into 32 patients with MIBC at UCSF. They are visible with any imaging modality, including all on- table imaging. They are safe and easy to place, appear not to migrate, and are durable. We propose that our fiducial markers serve to increase high-dose radiotherapy targeting accuracy and lower collateral radiation. The goals of the work proposed here are to (1.) Assess the efficacy, treatment benefit, and generalizability of our markers and placement technique among a larger sample of patients with MIBC; 2. Use our fiducial markers to assess the degree to which urologists may miss occult tumor outside of the main resection area; and 3. Use our markers to assess the positive and negative predictive value (PPV, NPV) of diffusion-MRI to detect occult tumor pre and post-treatment. To accomplish these goals, we will perform 2 separate but complementary studies at UCSF--one in collaboration with Harvard/MGH.